Evidence Conflict

Follow this topic Ask about this topic

Momentum

7 papers in the last four weeks, up 133% on the four weeks before. 0.1% of all new papers.

Jul 6Week of Sep 21

Latest papers 39

All topics

When Do Biological Reasoning Models Use Their Biological Inputs?
Oct 1, 2026Ada Fang, Nikitha Thoduguli, Lukas Fesser +3Evidence Conflict Foundation Model
Harvard University · Massachusetts Institute of Technology
Can LLMs Value the Right Evidence? Evidence-Value Misalignment in Dynamic Medical Diagnosis
Sep 28, 2026Kehua Feng, Yunsheng Lu, Yitong Qiao +5Diagnosis Evidence Conflict
Zhejiang University · Ant Healthcare, Ant Group
ReSync: Re-Aligning the Two Clocks of Asynchronous World-Action Models
Sep 27, 2026Xi Lin, Feihong Zhang, Yulong Shi +6Recent World-Action Models World Models
Johns Hopkins University · Tsinghua University · Yinwang Intelligent Technology Co., Ltd. +1
ARAFA: An LLM-Generated Arabic Fact-Checking Dataset
Sep 22, 2026Christophe Khalil, Shady Elbassuoni, Rida AssafArabic Arabic Natural Language Processing
Computer Science, American University of Beirut, Beirut, Lebanon
GroundedGEO: Auditing the Evidence Gap in Generative Search Rankings
Sep 21, 2026Yihan Xia, Huiling Fan, Kangrong Zhong +1Evidence-Grounded Questions Evidence Conflict
Shenzhen University Shenzhen, China
Look Where It Counts: A Free, Label-Free Visual Evidence Signal for Fine-Grained Vision-Language Reasoning
Sep 21, 2026Santi Ram Tiwari, Nihal Naik, Devbrat Pandey +1Fine-Grained Perception Recent Vision-Language Models
KGraph AI Solutions Pvt. Ltd. Bangalore, India – 560016
When Evidence Conflicts: Reliability-aware Meta-review Generation
Sep 21, 2026Xinzhe Wang, Fei Tao, Jiang Xie +2Review Evidence Conflict
Chongqing University of Posts and Telecommunications, China · NewsBreak, USA
Beyond Accuracy: How Procedural Traces Shift the Decision Criterion of LLM Overseers
Sep 16, 2026Zihan Chen, Di Zhu, Lei Zheng +1Model Auditing Large Language Models Fail
Stevens Institute of Technology Hoboken, NJ, USA · University of Massachusetts Boston Boston, MA, USA · Stony Brook University Stony Brook, NY, USA
Can LLMs Follow Medical Expert Logic? A Benchmark for Hierarchical Logical Consistency in Risk-of-Bias Assessment
Sep 12, 2026Jiayu Huang, Zichen Tang, Qianhui Ling +2Clinical Reasoning Training Large Language Model Reliability
Beijing University of Posts and Telecommunications · Hypertension Center, Beijing Anzhen Hospital, Capital Medical University
Every pooling rule has its world: matching probability combination rules to situations and stakes
Aug 11, 2026Tanel Tammet, Priit Järv, Dirk DraheimProbability Probabilistic Model
Tallinn University of Technology
EvoTrustRAG: Evolution-Aware Conflict Attribution and Evidence Handling for Reliable Retrieval-Augmented Generation
Aug 8, 2026Xi Nie, Hongwei Li, Shenghao Wu +3Evidence Conflict
School of Computer Science and Engineering, University of Electronic Science and Technology of China · Institute of Information Engineering, Chinese Academy of Sciences
Artificial Intelligence Can Match Domain Experts in Evidence Extraction and Critical Appraisal of Microbial Oncogenesis Research Publications
Aug 7, 2026Kaela Kokkas, Hairong Wang, Richard Klein +11Systematic Review Biomedical Text
Department of Clinical Microbiology and Infectious Diseases, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, South Africa · School of Computer Science and Applied Mathematics, University of the Witwatersrand, Johannesburg, South Africa · Wits Machine Intelligence and Neural Discovery (MIND) Institute, University of the Witwatersrand, Johannesburg, South Africa +13
Evidence-Based Scientific Question Discovery: A Framework with Historical Backtesting
Jul 29, 2026Hui MaoScientific Discovery Evidence-Grounded Questions
Independent Researcher
Same Evidence, Different Target: Decoding How Diagnostic Evidence Bears on Causal Questions from Language-Model States
Jul 29, 2026Weiyi Kong, Zhuoran LiDiagnosis Evidence Conflict
University of Toronto, Toronto, Canada · The University of Hong Kong, Hong Kong
Diagnosing Fine-Grained Inconsistency Classification in Financial Disclosure Text
Jul 29, 2026Aman Kumar, Lasitha Vidyaratne, Dipanjan D Ghosh +2Large Language Model Classification Evidence Conflict
Hitachi America, Ltd.
Auditing Evidence Use in Medical LLM Diagnosis
Jul 23, 2026Junchi Liao, Jiawen Deng, Fuji RenDiagnosis Evidence Conflict
University of Electronic Science and Technology of China Chengdu, China
Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows
Jul 12, 2026Keyur GabaniFraud Detection Large Language Model Reliability
The strength of clinical evidence is recoverable from language model representations but not from their stated grades
Jun 27, 2026Soroosh Tayebi ArastehLarge Language Model Reliability Evidence Conflict
Lab for AI in Medicine, RWTH Aachen University, Aachen, Germany · Department of Diagnostic and Interventional Radiology, University Hospital RWTH Aachen, Aachen · Department of Urology, Stanford University, Stanford, CA, USA +1
ConflictScore: Identifying and Measuring How Language Models Handle Conflicting Evidence
Jun 24, 2026Siyi Liu, Aaron Halfaker, Dan Roth +1Evidence Conflict Truth
University of Pennsylvania · Microsoft
T2D-Bench: Evidence-Gated Evaluation of LLM Outputs for Type 2 Diabetes Using a Multi-Layer Clinical-Lifestyle Knowledge Graph
Jun 23, 2026Saba A. Farahani, Hung Cao, Ramesh Jain +1Diabetes Biomedical Knowledge Graph
University of California, Irvine, Irvine, CA, USA
Where Is My Physics Wrong? Localized and Identifiable Discovery of Model Discrepancy
Jun 22, 2026Yifan WangDiscrepancy Model Discovery
Department of Mechanical Engineering, McGill University, QC, H3A 2T7, Canada
Visualizing Uncertainty: Spatial Maps of Missing and Conflicting Evidence in Deep Learning
Jun 14, 2026Dong Hyun Jeong, Feng Chen, Jin-Hee Cho +3Class Activation Mapping Pixel-Level Prediction
University of the District of Columbia Department of Computer Science Washington, DC, USA · University of Texas at Dallas Department of Computer Science Richardson, TX, USA · Virginia Tech Department of Computer Science Blacksburg, VA, USA +3
Multi-Agent Framework for Audit Risk Assessment with Explicit Uncertainty and Evidence Conflict Modeling
Jun 14, 2026Yuhan Wang, Manqing Wang, Yixuan Lu +2Model Auditing Evidence Conflict
Columbia University New York, USA · Trine University Phoenix, USA · University of Sofia Palo Alto, USA +2
X-MADAM-RAG: Diagnosing and Handling Chinese-English Evidence Conflict in Retrieval-Augmented Generation
Jun 11, 2026Yongqi Kang, Yu Fu, Yong ZhaoHievi-Rag Evidence Conflict
Sichuan University, Chengdu, China
AbstRAG: Learning to Abstract for Retrieval Problems
Jun 8, 2026Lei Xu, Xin Quan, Daniel Pedronette +1Question-Answering Benchmarks Evidence Conflict
Idiap Research Institute, Switzerland · São Paulo State University, Brazil · Department of Computer Science, University of Manchester, United Kingdom +1
Cherry-pick Override: Unsafe Directional Commitment in LLM Judges under Mixed Evidence
Jun 5, 2026Haoran XuVerdict Evidence Conflict
University of Glasgow
Evaluating RAG Reliability under Clean, Misleading, and Mixed Retrieval
Jun 5, 2026Sevgi Yigit-SertMisinformation Evidence Conflict
Computer Engineering Department, Ankara University, Golbasi Campus, 06830, Ankara, Turkey · Golbasi Campus, 06830, Ankara, Turkey
Output Type Before Quality: A Standards-Derived XAI Admissibility Rubric for Autonomous-Driving Safety
Jun 3, 2026Abhinaw Priyadershi, Mandar Pitale, Jelena Frtunikj +1Xai Autonomous Driving
NVIDIA Corporation, Santa Clara, CA, USA · NVIDIA GmbH, Munich, Germany
Encoded but Not Routed: Explaining the Table-Chart Gap in Scientific Claim Verification
Jun 1, 2026Sunisth Kumar, Xanh Ho, Tim Schopf +3Evidence Conflict Chart
The University of Tokyo · NII LLMC · National Institute of Informatics +2
Position: Anthropomorphic Misalignment Research Needs Stronger Evidence
May 29, 2026Vansh Gupta, Peter Nutter, Samuel Stante +5Anthropomorphism Evidence Conflict
1ETH Zurich
StepGap: A Hybrid NLI-LLM Checker for Step-Level Evidence-Gap Detectionin Multi-Hop Question Answering
May 23, 2026Yuelyu Ji, Zhuochun Li, Hui Ji +1Multi-Hop Reasoning Evidence Conflict
School of Computing and Information, University of Pittsburgh
CounterCount: A Diagnostic Framework for Counting Bias in Vision Language Models
May 18, 2026Reem Alzahrani, Hassan Alshanqiti, Bushra Bin Hemid +3Counting Multimodal Reasoning
King Abdullah University of Science and Technology, Thuwal, Saudi Arabia. · University of Edinburgh · KAUST
Position: State-of-the-Art Claims Require State-of-the-Art Evidence
May 17, 2026YongKyung OhArtificial Intelligence Benchmarks Evidence Conflict
Medical & Imaging Informatics (MII), University of California, Los Angeles (UCLA)
When Evidence Conflicts: Uncertainty and Order Effects in Retrieval-Augmented Biomedical Question Answering
May 13, 2026Yikun Han, Mengfei Lan, Halil KilicogluEvidence Conflict Biomedical Text
University of Illinois Urbana-Champaign
VideoSEAL: Mitigating Evidence Misalignment in Agentic Long Video Understanding by Decoupling Answer Authority
May 12, 2026Chenhao Qiu, Yechao Zhang, Xin Luo +2Long Video Question Answering Temporal Grounding
Mango TV · Mango TV, Changsha, China · Nanyang Technological University +1
ArbGraph: Conflict-Aware Evidence Arbitration for Reliable Long-Form Retrieval-Augmented Generation
Apr 20, 2026Qingying Niu, Yuhao Wang, Ruiyang Ren +2Evidence Conflict Arbitration
Gaoling School of Artificial Intelligence, Renmin University of China, Beijing, China, 100872 · Carnegie Mellon University, Pittsburgh, Pennsylvania, USA, 15213
CARE: Privacy-Compliant Agentic Reasoning with Evidence Discordance
Apr 1, 2026Haochen Liu, Weien Li, Rui Song +6Evidence Conflict Mimic
University of Cambridge · McGill University · MBZUAI - Mohamed bin Zayed University of Artificial Intelligence +2
Disentangling Reasoning Logic to Resolve Explicit Knowledge Conflicts
Aug 2, 2025Xianda Zheng, Zijian Huang, Meng-Fen Chiang +4LLM Reasoning Strategies Reasoning Skills
School of Computer Science, University of Auckland · Department of Electronics and Electrical Engineering, National Yang Ming Chiao Tung University · School of Computing and Information Systems, Singapore Management University +1
DrugAgent: Reliable Multi-Agent Integration of Conflicting Biomedical Evidence for Drug-Target Interaction Assessment
Aug 23, 2024Yoshitaka Inoue, Tianci Song, Xinling Wang +3Drug-Drug Interactions Drug Discovery
Department of Computer Science and Engineering, University of Minnesota, Minneapolis, MN, USA · Computational Biology Branch, National Library of Medicine, Bethesda, MD, USA · Khoury College of Computer Sciences, Northeastern University, Arlington, VA, USA +2

Oct 1, 2026, cs.LGJ/K move · Enter open · S save

When Do Biological Reasoning Models Use Their Biological Inputs?

Ada Fang, Nikitha Thoduguli, Lukas Fesser, Hanlin Zhang+2

Harvard University · Massachusetts Institute of Technology

Evidence Conflict Foundation Model Intermediate Checkpoints

Open paper AskPDF arXiv

Biological reasoning models use post-training to connect LLMs to biological foundation model representations and biological text. Their benchmark accuracy is taken as evidence that LLMs reason over these inputs. We test this assumption in six biological reasoning models across DNA, protein, and single-cell tasks. We perturb one biological input while holding the query and other inputs fixed, construct evidence conflicts that pair the foundation model representation of one genome, protein, or cell with the text of another, fit linear probes to the representations the language model receives, and analyze reasoning traces against the biological inputs. Evo2 and ESM3 contribute little to BioReason and BioReason-Pro performance on the evaluated tasks. Shuffling the DNA sequence barely changes BioReason disease prediction accuracy, and in evidence conflicts the two models follow the text in 97.9% and 99.7% of cases. Linear probes trained on the Evo2 and ESM3 representations predict the task targets, so these foundation models encode information relevant to the task, but provide limited overall performance improvement to BioReason and BioReason-Pro. In contrast, foundation model inputs contribute to ChatNT, Prot2Text-V2, and CellWhisperer performance, and differentially expressed genes in the gene sentence contribute to Cell2Sentence-Scale performance. Across SFT and RL checkpoints of BioReason-Pro and 42 BioReason checkpoints, increases in accuracy do not imply greater performance contributions from biological inputs. BioReason traces misstate nucleotide changes, while BioReason-Pro traces describe functions omitted from final predictions under evidence conflicts. We find that current post-training strategies do not ensure that foundation model representations contribute to task performance.

Momentum

Related topics

Latest papers 39